Gsd-skill-creator phonics-decoding
Letter-sound relationships, decoding strategies, and word-attack skills for reading unfamiliar text. Covers alphabetic principle, phoneme-grapheme correspondences (single letters, digraphs, diphthongs, r-controlled vowels), syllable types (closed, open, VCe, vowel team, r-controlled, consonant-le), morphemic analysis (prefixes, suffixes, roots), multisyllabic word strategies, and fluency development through automaticity. Use when teaching decoding, diagnosing reading errors, analyzing miscue patterns, or building word-attack routines.
git clone https://github.com/Tibsfox/gsd-skill-creator
T=$(mktemp -d) && git clone --depth=1 https://github.com/Tibsfox/gsd-skill-creator "$T" && mkdir -p ~/.claude/skills && cp -r "$T/examples/skills/reading/phonics-decoding" ~/.claude/skills/tibsfox-gsd-skill-creator-phonics-decoding && rm -rf "$T"
examples/skills/reading/phonics-decoding/SKILL.mdPhonics & Decoding
Decoding is the process of translating written symbols into spoken language. It is the mechanical foundation on which all reading comprehension rests. A reader who cannot decode accurately and automatically will exhaust cognitive resources on word identification, leaving nothing for meaning-making. This skill covers the alphabetic principle, phoneme-grapheme mappings, syllable types, morphemic analysis, and the path from labored decoding to automatic word recognition.
Agent affinity: chomsky-r (language structure, syntax-phonology interface), clay (Running Records, miscue analysis, early literacy)
Concept IDs: read-phonological-awareness, read-phonics-decoding, read-sight-words, read-reading-fluency
The Alphabetic Principle
English is an alphabetic writing system: written symbols (graphemes) represent spoken sounds (phonemes). The alphabetic principle is the insight that letters and letter combinations map to sounds in a systematic, learnable way. This principle is not obvious -- Chinese uses logographic writing, Japanese uses syllabaries, and even English's spelling-to-sound correspondences are famously irregular. But the system is far more regular than its reputation suggests: approximately 84% of English words follow predictable phonics patterns (Hanna et al., 1966).
Why it matters for reading agents. When a reader encounters an unfamiliar word, they need a strategy. The alphabetic principle provides the primary strategy: sound it out using known letter-sound correspondences, then check whether the resulting pronunciation matches a word in the reader's oral vocabulary.
Phoneme-Grapheme Correspondences
Consonants
Most consonant letters map to a single phoneme reliably: b, d, f, h, j, k, l, m, n, p, r, t, v, w, y, z. The exceptions are:
| Grapheme | Phonemes | Rule |
|---|---|---|
| c | /k/ before a, o, u; /s/ before e, i, y | "Soft c" rule |
| g | /g/ before a, o, u; /j/ before e, i, y (with exceptions) | "Soft g" rule (exceptions: get, give, girl) |
| s | /s/ at word start; /z/ between vowels or word-final after voiced sounds | Voicing assimilation |
| x | /ks/ typically; /gz/ in unstressed initial syllables (exact, exist) | Stress-dependent |
Consonant Digraphs
Two letters representing a single phoneme distinct from either letter alone:
| Digraph | Phoneme | Examples |
|---|---|---|
| sh | /sh/ | ship, wash, mission |
| ch | /ch/ | chip, lunch, match |
| th | /th/ (voiced) or /th/ (voiceless) | this (voiced), thin (voiceless) |
| wh | /w/ (most dialects) | when, where, which |
| ph | /f/ | phone, graph |
| ck | /k/ | back, duck (after short vowel) |
| ng | /ng/ | ring, song |
Vowels -- Short and Long
English has five vowel letters representing at least 15 distinct vowel phonemes. The short/long distinction is fundamental:
| Vowel | Short (CVC pattern) | Long (CVCe or open syllable) |
|---|---|---|
| a | /a/ as in cat | /ay/ as in cake, ba-by |
| e | /e/ as in bed | /ee/ as in Pete, me |
| i | /i/ as in sit | /eye/ as in kite, hi |
| o | /o/ as in hot | /oh/ as in bone, go |
| u | /u/ as in cup | /yoo/ as in cute, mu-sic |
Vowel Teams and Diphthongs
| Pattern | Sound | Examples | Memory aid |
|---|---|---|---|
| ai, ay | /ay/ | rain, play | "When two vowels go walking, the first one does the talking" |
| ee, ea | /ee/ | tree, read | (ea also says /e/: bread, head) |
| oa, ow | /oh/ | boat, snow | (ow also says /ow/: cow, now) |
| oi, oy | /oy/ | coin, boy | Diphthong -- glides from one vowel to another |
| ou, ow | /ow/ | cloud, cow | Diphthong |
| oo | /oo/ or /oo/ | moon (long), book (short) | Two pronunciations |
| au, aw | /aw/ | cause, saw |
R-Controlled Vowels
When a vowel is followed by r, the vowel sound changes:
| Pattern | Sound | Examples |
|---|---|---|
| ar | /ar/ | car, star |
| er, ir, ur | /er/ (all three merge) | her, bird, burn |
| or | /or/ | for, corn |
This merger of er/ir/ur is one of the most common spelling challenges in English: the same sound is spelled three different ways with no reliable rule for choosing among them.
Six Syllable Types
Every English syllable falls into one of six types. Knowing the type predicts the vowel sound, making syllable-type identification the most powerful decoding strategy for multisyllabic words.
| Type | Pattern | Vowel sound | Examples |
|---|---|---|---|
| Closed | Ends in consonant | Short | cat, rab-bit, pump-kin |
| Open | Ends in vowel | Long | me, ba-by, pi-lot |
| VCe (Magic e) | Vowel-consonant-e | Long | cake, com-pete, in-vite |
| Vowel Team | Two vowels together | Varies by team | rain, boat, coin |
| R-Controlled | Vowel + r | R-modified | car, her, corn |
| Consonant-le | C + le (final syllable) | Schwa + l | ta-ble, puz-zle, sim-ple |
Decoding strategy. When encountering a multisyllabic word: (1) divide into syllables, (2) identify the type of each syllable, (3) apply the vowel rule for that type, (4) blend the syllables, (5) check against oral vocabulary.
Syllable Division Rules
To decode a multisyllabic word, the reader must first divide it into syllables. Four patterns cover most cases:
| Rule | Pattern | Division | Example |
|---|---|---|---|
| VC/CV | Two consonants between vowels | Split between consonants | rab/bit, nap/kin |
| V/CV | One consonant between vowels (first try) | Split before consonant | pi/lot, ba/by |
| VC/V | One consonant between vowels (if V/CV fails) | Split after consonant | riv/er, cab/in |
| V/V | Two vowels not forming a team | Split between vowels | cre/ate, po/em |
Decision procedure. Try V/CV first (it produces an open first syllable with a long vowel). If the resulting pronunciation does not match a known word, try VC/V (closed first syllable, short vowel). This flexible strategy allows self-correction.
Morphemic Analysis
Beyond phonics, readers decode words by recognizing meaningful units (morphemes): prefixes, suffixes, and roots.
High-Frequency Prefixes
The four most common prefixes account for 58% of all prefixed English words (White, Sowell, & Yanagihara, 1989):
| Prefix | Meaning | Examples |
|---|---|---|
| un- | not, opposite | unhappy, undo |
| re- | again, back | rewrite, return |
| in-/im-/il-/ir- | not | invisible, impossible |
| dis- | not, opposite | disagree, disconnect |
High-Frequency Suffixes
| Suffix | Function | Examples |
|---|---|---|
| -s, -es | Plural / third person | cats, wishes |
| -ed | Past tense | walked, jumped |
| -ing | Present participle | running, reading |
| -ly | Adverb | quickly, slowly |
| -tion, -sion | Noun (from verb) | creation, decision |
| -ful | Full of | hopeful, careful |
| -less | Without | hopeless, careless |
| -ment | Noun (from verb) | movement, enjoyment |
Latin and Greek Roots
For academic and technical vocabulary, root knowledge is the most powerful decoding tool:
| Root | Origin | Meaning | Examples |
|---|---|---|---|
| dict | Latin | say, speak | predict, dictate, verdict |
| struct | Latin | build | construct, instruct, structure |
| graph/gram | Greek | write | paragraph, telegram, biography |
| port | Latin | carry | transport, import, portable |
| spect | Latin | look | inspect, spectacle, perspective |
| aud | Latin | hear | audience, audio, auditorium |
| bio | Greek | life | biology, biography, antibiotic |
| chron | Greek | time | chronology, synchronize |
From Decoding to Automaticity
Decoding is necessary but not sufficient. The goal is automaticity -- word recognition so fast and effortless that attention is fully available for comprehension. Ehri (2005) describes four phases of word-reading development:
- Pre-alphabetic: Recognizes words by visual features (the golden arches = McDonald's), not letter-sound mappings.
- Partial alphabetic: Uses some letter-sound cues, typically first and last letters.
- Full alphabetic: Applies complete phoneme-grapheme knowledge to decode unfamiliar words.
- Consolidated alphabetic: Recognizes multi-letter patterns (morphemes, syllables, rimes) as units, achieving rapid recognition.
The transition from Phase 3 to Phase 4 is where fluency emerges. Practice with connected text drives this transition -- isolated word drills are necessary but not sufficient.
Running Records and Miscue Analysis
When a reader makes an error (miscue), the type of error reveals which decoding skills need attention:
| Miscue type | What the reader did | Skill gap |
|---|---|---|
| Visual substitution | Said "house" for "horse" | Not attending to all letters -- partial alphabetic |
| Phonetic substitution | Said "beg" for "big" | Vowel confusion -- needs vowel discrimination work |
| Nonsense word | Said "blunk" for "blank" | Applying phonics but not cross-checking with meaning |
| Omission | Skipped "unfortunately" | Overwhelmed by multisyllabic word -- needs syllable strategies |
| Insertion | Added a word not in the text | Over-relying on prediction, under-relying on print |
Clay's Running Records (1993) provide a systematic notation for recording these miscues during oral reading. The pattern of errors, not the error count, drives instructional decisions.
When to Use This Skill
- Teaching or diagnosing letter-sound relationships
- Analyzing reading errors (miscues) for instructional planning
- Building word-attack strategies for unfamiliar words
- Developing automaticity and fluency
- Morphemic analysis of academic vocabulary
When NOT to Use This Skill
- For meaning-making and comprehension strategies -- use reading-comprehension
- For vocabulary instruction beyond decoding -- use vocabulary-development
- For critical analysis of text quality or argument -- use critical-reading
- For literary interpretation -- use literary-analysis
- For evaluating sources -- use information-literacy
Cross-References
- chomsky-r agent: Language structure and syntax awareness. Chomsky's generative grammar illuminates why English spelling is "deeper" than surface phonetics -- it preserves morphological relationships (sign/signal, bomb/bombard) at the cost of phonetic transparency.
- clay agent: Running Records, Reading Recovery, early literacy assessment. Primary agent for diagnosing decoding difficulties and planning intervention.
- vocabulary-development skill: Where decoding meets meaning -- once a word is decoded, vocabulary knowledge determines whether it is understood.
- reading-comprehension skill: The purpose of decoding. Automaticity in decoding frees cognitive resources for comprehension.
References
- Adams, M. J. (1990). Beginning to Read: Thinking and Learning about Print. MIT Press.
- Clay, M. M. (1993). An Observation Survey of Early Literacy Achievement. Heinemann.
- Ehri, L. C. (2005). Learning to read words: Theory, findings, and issues. Scientific Studies of Reading, 9(2), 167-188.
- Hanna, P. R., Hanna, J. S., Hodges, R. E., & Rudorf, E. H. (1966). Phoneme-Grapheme Correspondences as Cues to Spelling Improvement. U.S. Office of Education.
- Moats, L. C. (2020). Speech to Print: Language Essentials for Teachers. 3rd edition. Brookes Publishing.
- White, T. G., Sowell, J., & Yanagihara, A. (1989). Teaching elementary students to use word-part clues. The Reading Teacher, 42(4), 302-308.